*2.4. Constrained Scenarios*

Base scenarios and optimization scenarios are designed to fully simulate the future PLE space layout of Zhaotong city in 2030. The optimization scenarios include Scenario A, in which production and living development are given priority; Scenario B, in which ecological protection is given priority; and Scenario C, in which both are considered. In addition, three levels (high level, medium level, and low level) are designed in every optimization scenario. The details are shown in Table 6.

**Table 6.** Scenario design and solution code.


There are two main aspects in which the above scenarios differ. On the one hand, the control variables' values are different in the system dynamics model for quantitative optimization. On the other hand, the transformation rules are different when using the FLUS model for spatial simulation.

Five different approaches are chosen as optimization projects in the PLE space system dynamics model, including land-use efficiency promotion, industrial structure adjustment, agricultural production space protection, intensive development of construction land, and controlling population growth. Relevant control variables in the model are listed in Table 7. The variables associated with the land-use efficiency promotion are mainly related to the GDP output per land (including GDPPA, GDPPI, and GDPLU). In the base scenario, i.e., at the previous rate of development, the values are 0.05, 11.58, and 9.10, respectively, by 2030. Based on the expert experience, on this basis, the optimization is carried out assuming that at high, medium, and low levels; GDPPA increases by 100%, 60%, and 30%, respectively; GDPPI increases by 60%, 30%, and 10%, respectively, and GDPLU increases by 120%, 80%, and 50% respectively. The values obtained are presented in Table 7. The values of the other variables for the different scenarios are also listed in Table 7.

**Table 7.** Base scenario and parameter settings at different levels.


<sup>1</sup> Definition of variables can be found in Table S1.

The cost matrix of the base scenario is simple; that is, the urban living land use cannot be converted into another land use, and the conversion between other different land types is unrestricted. See Table S7 for the detailed cost matrix. No masking of the restricted area is performed.

The development scenario of giving production-living development priority (Scenario A) ensures that the improvement of production space and living space is fully considered. The expansion of production-living space comes at the expense of occupying ecological space. In the stage of spatial simulation, urban living space cannot be converted

into others and is the same as priority agricultural production space and general agricultural production space (see Tables S8 and S9 for the detailed cost matrix). At the high and medium levels, the restricted ecological space vector scope of 2018 is used as a mask area; that is, the land parcels within the mask scope are no longer involved in the subsequent land-use conversion process. However, there is no mask at low levels.

The ecological space is fully protected in the context of Scenario B. For spatial simulation, urban living space cannot be converted into other land use and is the same as priority ecological space and general ecological space (see Tables S10 and S11 for the detailed cost matrix). Other settings are like Scenario A.

A balanced development scenario (Scenario C) means that the priority ecological space and priority agricultural production space are fully protected. During the spatial simulation, urban living space cannot be converted into another land use, and the same requirements are made for the priority ecological space and priority agricultural production space (see Tables S12 and S13 for the detailed cost matrix). Other settings are similar to Scenario A.

Only the industrial production space area, urban living space area, and rural living space area can be directly obtained from the simulation results of the system dynamic model. Other PLE space needs to be calculated according to the proportion of subclasses to the upper-class land in 2018, especially in the base scenario and Scenario B, in which ecological space protection is given priority. In Scenario A, the areas of priority agricultural production space and general agricultural production space were the same as those in 2018, and others were obtained by the above method. In Scenario C, the priority ecological space and priority agricultural production space were not less than those in 2018, and others were obtained by the above method.
